Current Works in Mineral Processing, Vol. 1, No.

1, March, 2019
30 https://dx.doi.org/10.22606/cwimp.2019.11004

Review on the Applications of Apparent Mean Shape Factor on

the Integration of Coarse and Fine PSDs Measured by Different
Techniques: Quartz Example

U. Ulusoy1,*, M. Yekeler1, O.Y. Gülsoy2, N. A. Aydoğan2, C. Biçer1 and, Z. Gülsoy1

1
Department of Mining Engineering, Sivas Cumhuriyet University, TR-58140, Sivas, Turkey
2
Department of Mining Engineering, Hacettepe University, TR-06532, Ankara, Turkey
Email: uulusoy@cumhuriyet.edu.tr

Abstract. In most industry where grinding is utilized, particle size, which is a decisive element in
establishing the productivity of production processes and execution of the end product, is analyzed to
describe the size distribution of particles in a given sample. In mineral and coal processing, particle
size distributions (PSDs) of particulate materials were traditionally accomplished by sieving, which
gives inaccurate particle size and PSD below 38 µm. This paper reviews the studies related to the
combinations PSDs of different mill products of the same quartz mineral by using different particle
size measurement techniques to build the whole PSDs including coarse and fine PSDs. For this
purpose, almost pure quartz mineral (which is the most suitable brittle material that gives first order
grinding kinetics) ground by ball and rod mill products that are the most widely used conventional
mills in mineral processing were measured by different size analysis techniques, i.e. sieving for coarse
sizes, Andreasen pipette sedimentation, and laser diffraction for fine sizes below 38 µm and combined
them to construct whole size distribution by using apparent mean shape factor, r.
The results were satisfactorily well for both cases; PSDs by laser diffraction size distribution-
sieving and PSDs by Andreasen pipette sedimentation-sieving, i.e., a smooth overlap of corrected
laser diffraction and sieving PSDs and Andreasen pipette sedimentation and sieving PSDs were
obtained by applying to the particle size distribution with r shifting to the right side of the curves. In
the case of determination of PSDs by laser diffraction and sieving, r values determined from the
corrected particle size distributions were found to be 1.29 and 1.25 for ball and rod milled products,
respectively. The results indicates that there is not significant differences between the shape factors
of ball and rod milled products of quartz mineral, i.e. both of them have irregular particles, which
deviates from the spherical shape as proved by their SEM microphotographs. On the other hand, for
the PSDs by Andreasen pipette sedimentation and sieving, the corrected sedimentation data came
closer to the sieving data. It was found that r values determined from the corrected PSDs of the same
quartz mineral ground by ball and rod mill were 1.00 and 1.12, respectively. The results show that
the rod milled products were not more regular in shape than ball milled products as evidenced by
SEM pictures and previous works. Thus, this approach can be utilized for the integration of PSDs
analyzed by different techniques for coarse and fine sizes of fine particulate coals, minerals, and
similar materials ground finely.

Keywords: Quartz, sedimentation size, sieve size, laser diffraction size, particle size distribution
(PSD), apparent mean shape factor.

Nomenclature
d cm Sieving size or particle diameter, the size of the largest particle
that can still be existing in the sample
D mm inner diameter of mill or particle size measured by any particle size
measurement technique
d50 µm 50% (by mass) passing size
Df g/cm3 density of fluid
DS g/cm3 density of solid particle
fC -- fractional powder filling
g cm/sec.2 acceleration due to gravity

CWiMP Copyright © 2019 Isaac Scientific Publishing

Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019 31

H cm height in sedimentation
JB % fractional ball filling
L mm mill length
Nc rpm critical speed
PSD -- particle size distribution
r -- apparent mean shape factor
t sec. settling time
U -- powder-ball loading ratio
v cm/sec. velocity
V cm3 mill volume
W0 g weight of initial sample
Wd g weight of particle size d
 centipoise (g/cm.sec.) viscosity of the fluid

1 Introduction
A particle is a separate part of material irrespective of its size. Particle size, which is a term used to
compare dimensions of mostly solid particle, is a critical parameter for a variety of operations in
materials science and technology as well as many other branches such as medicine, pharmacology,
biology, ecology, energy technology and the geosciences [1].
The behavior and characteristics of particulate material are mainly dependent on their size,
morphology and size distributions since the size impacts get progressively crucial, as its size gradually
reduced [2]. Precise size measurement is of essential since quality and performance of several particulate
products are primarily depend on size distribution of fine particles. So particle size distribution or
shortly PSD can be defined as the size and number of observations of particles in a particulate system
or group. It influences the attributes of a powder and dispersions in many crucial manner [3] and it is
generally presented as percentage of the entire mass of particle hold by a given size fraction. It is
significant to determine and check the PSD of particulate materials because particle size is a key index
of quality and performance in various industries related to powders such as paint, pigment, cocoa
powder used in chocolate, the glass beads used in highway paint, cement and metal particles,
pharmaceutical powders, nano-particles, proteins, cosmetics, polymers, soils, abrasives, fertilizers, and
etc., [4]. Blending of metal or ceramic powders for producing sintered engineering components can be
given as a specific example why controlling and measuring PSD is important [5].
In mineral processing a spectrum of particle sizes (distribution) of ores are produced by size reduction
namely crushing and grinding. Particle size is one of the criteria for a product, which has to meet a size
specification for industrial minerals. The particle size of the product stream from the mill circuit acts a
significant part on the efficiency of the milling operations and consequently the whole plant throughput
since it is closely related to the recovery of valuable mineral [6]. In addition, particle size analysis of mill
products is important for providing the required liberation of the valuable minerals from the gangue
minerals at various particle sizes after grinding and also to enlarge the surface area for accelerating the
dissolution [7]. Above all, it is crucial to determine the optimum particle size of the separator feed in the
concentration circuits for maximum efficiency by detecting the particle size range at which any
separation losses occurred in the plant for the minimization since separation processes mainly depend on
the product particle size, which influences the efficiency of the separator [8]. Consequently, exact and
honest measurement of particle size and PSD are fundamental for the characterization of ores and
minerals. In coal processing technology, particle size and PSD have remarkable effects on wetting [9],
flotation recovery [10], flotation kinetics [11] and rheology of coal-water slurries [12]. The proper PSD of
the coal is required for an efficient combustion of pulverized carbon as PSD determines the performance
of the fluidized bed coal combustor [13]. In the mining industry of coal and ores as well as several other
fields using size analysis, the sieve size distribution is the common definition of particle size because
laboratory sieving gives good results with great accuracy for all particle sizes larger than 38 m when
good procedures and the correct equipment is used for the measurement. Below 38 m, the surface
energy of particles gets substantial, and the measurements are affected by agglomeration of particles
because of van der Waals forces. This makes particles retained on the sieve even when they are smaller

Copyright © 2019 Isaac Scientific Publishing CWiMP

32 Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019

than the sieve opening. Therefore, slight errors frequently become obvious below 38 m along with the
cumulative PSD displaying a small alteration of slope because of the extra particles accumulated below
the sieves smaller than 38 m [14].
But, PSD analysis must be exact and reliable, since important alterations or modifications in
processing plant operations can be made according to the laboratory test results. Therefore controlling
PSD is important for maximum efficiency and quality of the products [15, 16]. PSD analysis have
commonly utilized with accuracy for spherical particles. However, in practice particles are mostly
irregular and size characterization should also contain knowledge about shape of irregular particles. In
addition, their characterization in the course of analysis is strongly related to the method of
measurement utilized, therefore PSD determination is more significant labor due to the variability of the
shape and density of particles [17, 18] for understanding their physical and chemical attributes. Different
techniques of particle size measurement give different results of PSDs but conversion from one to
another is possible only when the shape factor is constant since usually the shape is a function of
particle size [19].
Production of micronized quartz particles is mainly required for mineral fillers in various industries
such as glass, paint, abrasives, refractory, and precision instruments. The suitability of fine quartz
particles for different industrial applications is determined by PSD. For example PSD is one of the
important properties of minerals used in paint [20]. Therefore, this paper reviews the studies [21, 22]
related to the combinations PSDs of different mill products of the same quartz mineral by using
different particle size measurement techniques to build the whole PSDs including coarse and fine PSDs.
For this purpose almost pure quartz mineral since this material had been used in many previous tests,
proved to be very consistent in grinding and sieving and also exhibits more brittle behavior for first
order breakage kinetics were used to grind in ball and rod mill, which are the most widely used
conventional mills. Then, fine and coarse PSDs of ball and rod mill products were measured by different
size analysis techniques (namely, sieving, Andreasen pipette sedimentation and laser diffraction) and
combined to construct whole PSDs by using apparent mean shape factor, r, which is also important for
the bringing out of the shape information to be extracted from the difference of the two PSDs.

2 Theoretical Background

2.1 Techniques of Particle Size Measurement

Sieving, which is one of the oldest methods [23, 24] of PSD analysis, is commonly used in mineral
processing because it is simple, cheap and regarded as a standard practice. It includes shaking the
sample by a sieve series stacked from the biggest to the smallest [3]. Then, calculation of the weight
percentage of each size fraction can be made by weighing the samples retained on each sieve when
passing a known amount of sample material in proper sequence through smaller sieves [8]. It gives a
mass distribution and PSD report as cumulative distribution. Sieving measures a size termed as the
sieve diameter, which is relied on the upper limit of the width and thickness terms of the particle since
the length of the particle does not impede its transition through the sieve opening except for highly rod-
like particles. Therefore, classification in sieving accomplished based on particle width. If it is performed
meticulously using standard procedure, it produces reliable and reproducible results about PSD [25] for
particles coarser than 38 m, but measurement for dry powders under 38 µm is not easy. Although wet
sieving is thought to overcome this problem the results generally yield very poor reproducibility [26].
Therefore, substantial information is also included in the PSD of ultrafine minerals below 38 m [27].
Sedimentation technique, which is versatile, because it can deal with any powder that can be
dispersed in a liquid, has also been used as a traditional method for the paint and ceramics industry in
the applicable range of 2-50 µm [28, 26]. Andreasen pipette, is a laboratory tool for sedimentation
technique to determine the velocity of particles settling in a fluid because of the gravitational forces
acting on the particle, versus the buoyancy of the fluid and other drag forces [3]. It allows a
homogeneous suspension to settle in a glass cylinder and gives size known as Stokes diameter, which is
simply a comparison of the particle’s settling rate to a sphere settling at the same rate. Even though it
is cheap, analysis is tedious and time-consuming [24]. PSD can be measured by the falling velocity of
particles suspended in a viscous liquid, i.e., each representative sample, which will have deposited under

CWiMP Copyright © 2019 Isaac Scientific Publishing

Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019 33

the sampling level is taken from the settling suspension at a constant horizontal level for a period of
time.
Laser diffraction, which is known for its operational simplicity gives robust and versatile solutions for
PSD analysis since today’s laboratory instruments are more flexible, easy to use and, more automated
[29] therefore, this technique is now become a popular and standard in the laboratory [30] for particle
sizes between 0.1 and 2 000 µm [31]. It is based on the assumption of diffraction angle, which is
inversely proportional to particle size, i.e. the observed angle of light scattering increments
logarithmically as particle size lessens. Besides, scattering intensity is also contingent on particle size,
decreasing with the volume of particle. Hence, big particles scatter light at lower angles with great
intensity, on the other hand small particles scatter at greater angles, but with small intensity [31]. The
data of angular scattering intensity is then analyzed to figure out the size of the particles responsible for
the formation of the scattering pattern based on the Mie theory [30] and described by the volume
equivalent sphere diameter as volume distribution, which is also the same as the weight distribution for
constant density.

2.2 Approach of Apparent Mean Shape Factor (r)

As stated above sieving is the most commonly accepted and applied PSD analysis method in the mining
and mineral industry and therefore a single PSD integrated coarse and fine size distribution extended to
sub-sieve sizes (below 38 µm) is frequently in demand. For this aim, the particle size measured by the
particle size measurement technique is multiplied by a well known factor named as the apparent mean
shape factor r, which is defined previously by Allen [2]: the instrument size must be multiplied to give
the sieving size at which the same mass percentage less than size is obtained as stated by the equation 1.
d
r= (1)
D
where D represents the particle size measured by any particle size measurement technique and d is
denoted by the particle size measured by sieving. It should be noted that, r is not the same for all
material but depends on the PSD being analyzed. That’s why it is called as an apparent mean shape
factor. In this review, r factor was utilized as the ratio of the 50 % (by mass) passing size (which is also
known as d50) to the geometric mean sieving size as defined by the equation 2 [32], which yields a
method of calculating the factor according to the calibration of an instrument by a 21/2 sieve size
fraction and helps to unite PSDs, which have the Schuhmann form of cumulative PSD [33]:
d50
r= (2)
 
1/2
53  38
In other words, r corresponds to the Schuhmann plots shifted the PSD curve by any particle size
measurement technique to the right (or sometimes left) by the multiplication of the r factor to get the
corresponding sieving PSD, i.e., r factor actually describes the shape differences between 21/2 sieve size
fractions of milled materials.
Because, it is very difficult to obtain PSDs of fine industrial particles by using sieving technique,
which is not so true for less than 38 m, the PSDs of fine particles by using different techniques, such as
sieving, Andreasen pipette sedimentation and laser diffraction can be easy and the entire PSD including
coarse and fine PSD can be integrated by the combination of the two PSDs of the material by different
sizing techniques, i.e., fine PSD of sedimentation or laser diffraction and coarse PSD of sieving. This
approach also reveals the shape information (apparent mean shape factor) from the difference of the two
PSDs for the same material by utilization of different sizing techniques.

3 Experimental

3.1 Materials
In this review almost pure quartz mineral, which includes 97.20 % SiO2 as illustrated by Table 1, ground
by ball and rod mill (which are the most widely used conventional mills in mineral processing) products
were used [21, 22] to combine PSDs of them as illustrated in XRD pattern and SEM output by Figure 1.

Copyright © 2019 Isaac Scientific Publishing CWiMP

34 Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019

The specific gravity of quartz mineral used in this work was determined by the pycnometer test as 2.65
g/cm3.

Table 1. The chemical compositions of quartz minerals used in this study (modified from [21, 22].

Compound %
SiO2 97.20
Fe2O3 0.07
Al2O3 1.41
K2O 0.27
TiO2 0.06
CaO 0.20
MgO 0.08
Na2O 0.16

(a) (b)

Figure 1. Mineralogical analysis of the industrial minerals determined by techniques of (a) XRD (modified from
[22]) and (b) SEM

3.2 Grinding Tests

The grinding tests were used as the same method of previously reported studies [34, 35] as summarized
in Table 2. Two conventional mills, namely laboratory ball (200 x184 mm) and rod mills (200 x 280 mm)
were performed in dry conditions using with 5 475 kg of stainless steel balls of 30 and 26 mm and 22 600
kg of stainless steel rods of 29, 24 and 19 mm in size, respectively. 16 minutes of grinding time was used
in this work for both milling. While for ball milling, feed size fraction was –800+600 µm with a weight
of 367.3 g along with a ball filling of 20 % of the mill volume filled by the ball bed, –4.75+3.35 mm of
feed size fraction was used with a weight of 745.5 g for rod milling.

3.3 PSD Analysis by Different Size Analysis Techniques (Sieving, Laser Diffraction and
Andreasen Pipette Sedimentation)
After grinding the same quartz samples for 16 minutes, a representative sample by cone and quartering
technique was first wet sieved and then dry sieved for 10 minutes in a Ro-tap shaker without excessive
abrasion from a sieving kinetics study using Retsch Test sieve series comprise 4750 mm, 3350 mm, 800,
600, 425, 250, 150, 106, 75, 53 and 38 m for rod milling, while 800, 600, 425, 250, 150, 106, 75, 53 and
38 m sieves were used for ball milling. Grab sampling method was used for the samples of -38 m and -
53+38 m powder for the Andreasen pipette sedimentation and laser diffraction particle size analysis
tests.
The sedimentation tests were accomplished by using an Andreasen pipette (Figure 2), which includes
of a 200 mm graduated cylindrical flask containing an initially homogenous suspension and a pipette

CWiMP Copyright © 2019 Isaac Scientific Publishing

Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019 35

linked to a 10 ml reservoir via a two-way stopcock [21].

Table 2. Grinding conditions of the minerals studied (modified from [21]).

Type of milling Elements Parameters Values

Ball Milling Length x Inner diameter (L x D), mm 184 x 200
Mill Critical speed Nc, rpm 102
Operational speed (75% of critical speed), rpm 76
Type Stainless steel
Diameter, mm 26, 30
Average ball weights, g 68.19, 118.18
Balls
Specific gravity, g/cm3 7.90
Fractional ball filling, JB 0.2
Total ball weights, kg 5 475
Specific gravity, g/cm3 2.65
Total powder weight, g 367.3
Quartz Grinding time, min. 16
sample Fractional powder filling, fc 0.04
Powder-ball loading ratio, U 0.5
Feed size, µm -800+600
Rod Milling Length x Inner diameter (L x D), mm 280 x 200
Mill Critical speed Nc, rpm 102
Operational speed (50% of critical speed), rpm 51
Type Stainless steel
Diameter, mm 19, 24, 29
Rod Average rod weights, g 669, 1041, 1521
Specific gravity, g/cm3 7.90
Total rod weights, kg 22 600
Specific gravity, g/cm3 2.65
Quartz Total powder weight, g 745.5
sample Grinding time, min. 16
Feed size, µm -4 750+3 350

Figure 2. Andreasen pipette apparatus used in this experimental works for the sedimentation tests [21].

Since there is a certain particle size, d, which represents the size of the largest particle that can still
be existing in the sample equating to each settling length, H and time, t, Stokes diameter distribution of
the powder are calculated from Stokes' law for the various sampling times [8] as given in equation 3.

Copyright © 2019 Isaac Scientific Publishing CWiMP

36 Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019

The sedimentation procedure is followed by taking fixed volume of the sample at certain height, H
under the suspension surface after a sedimentation time, t. [36].
d 2g(Ds  D f )
  (3)
18 
where v is velocity, cm/sec.; d is particle diameter, cm; g is the gravity acceleration, cm/sec2; DS is
density of solid particle, g/cm3; Df is density of fluid, g/cm3 and  is viscosity, centipoises (g/cm.sec).
The solid concentration of the suspension in the pipette is about 1 % by weight with distilled water
having a viscosity of 0.0113 cp (g/cm.sec) at room temperature (20 ºC). Once the pipette was agitated 3
times by turning it upside down to make the suspension homogenous, an initial sample (W0) was taken
by applying suction to the top of the reservoir (10 ml volume) instantly by adjusting the two-way
stopcock at zero time (t=0) when the pipette was put into a smooth place, afterwards the stopcock was
inverted and letting the sample to pour into the glass container. The current suspension level was
recorded after each sample was drawn. Then, next samples were drawn based on the calculated settling
times, t (Table 3) of particles having 30, 25, 20, 15, 10 and 5 m size using the equation 4. The time
required for particles to settle in the Andreasen pipette of height H can be expressed as:
18H 
t 2 (4)
d g(Ds  D f )

Table 3. Settling times calculated from the Stokes’ equation for the -38 m fractions (modified from [21])

Particle size (m) Settling time (min.)

30 4 min. 25 sec.
25 6 min. 15 sec.
20 9 min. 33 sec.
15 16 min. 40 sec.
10 36 min. 35 sec.
5 146 min. 40 sec.

PSD determination of 21/2 sieve size fractions (-53+38 m) by Andreasen pipette was accomplished by
using the same routine of sedimentation test, which was repeated for particles having sizes of 53, 45, 38,
30 m according to the new settling times calculated by the equation 2. Settling times calculated from
the Stokes’ equation for the -53+38 m fractions were given in Table 4.

Table 4. Settling times calculated from the Stokes’ equation for the -53+38 m fractions of (modified from [21]).

Particle size (m) Settling time (min.)

53 1 min. 25 sec.
45 1 min. 56 sec.
38 2 min. 39 sec.
30 4 min. 10 sec.

At the end of the sedimentation test, all samples collected were dried in the oven at 75 ºC and
afterwards determined their weights using a balance (readability of 0.0001 g) to form the cumulative
PSD, which gives the fraction of the original material having a particle size less than d, using the ratio
of weight of sample taken to the corresponding initial weight [8] as in the equation 5.
W
Cumulative weight % passing  d  100 (5)
W0
where Wd is weight of particle size d, W0 is weight of initial sample.
In the case of PSD analysis by laser diffraction the instrument of Sympatec Helos, (H1305 model) [37],
was used. It is an optical instrument based on the measurement of laser light scattered by the particles
in the forward direction using Mie Theory.

CWiMP Copyright © 2019 Isaac Scientific Publishing

Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019 37

4 Results and Discussion

4.1 Integration of PSDs by Andreasen Pipette Sedimentation and Sieving Techniques

PSDs of quartz sample ground in ball and rod mills determined by sieving were given in Figure 3 (a).
By multiplying the value of cumulative weight % finer than 38 µm with PSDs measured by Andreasen
pipette sedimentation technique PSDs of quartz sample ground by ball and rod mills determined by
sedimentation method were plotted as in Fig. 3 (b). Although the PSDs are nearly the same in Figure 3
(a) the differences in the PSD curves were appeared in fine sections of the differently milled quartz
sample as seen from Figure 3 (b). This tells us some shape differences when the same mineral ground by
different mills.

(a) (b)

Figure 3. PSDs of quartz minerals ground in ball and rod mill by (a) Sieving and (b) Andreasen pipette
sedimentation techniques (Redrawn from [21])

Combining the coarse and fine PSD by sieving and sedimentation Figure 4 (a), which shows the
uncorrected PSDs of quartz mineral ground by different mills determined by the combination of two
different techniques was formed. Looking at the curves of whole PSD formed by this approach fine
section of the curves for rod milling product does not fit into the coarse section which was measured by
sieving, therefore this disagreement between coarse and fine PSDs measured by different techniques for
rod mill product can be used as a source of additional information on particle shape [38] but a good
trend was obtained by ball milling product indicating that there is no need to application of r factor, i.e,
the particle size measured by the sedimentation technique was well extended by the geometric mean
sieve size for the ball mill product. Thus, while the sedimentation size distribution should be shifted
right for the rod milled quartz product to obtain equivalent sieve size distribution, there is no need to
multiply the r factor for the ball milled quartz therefore the r value was assumed to be unity (r=1.00).
Then the d50 values were calculated as given in Table 5 from the sedimentation tests for the 21/2 sieve
size fraction (-53+38 m) of rod milled quartz sample, apparent mean shape factor, r, was ascertained
based on the equation 2. Afterwards, new particle sizes were determined by multiplying the particle size
by r factor, so that shifting to the right rod milled quartz, finally, corrected PSD curves were obtained
as depicted by Figure 4 (b), which confirms that there is good agreement between the sieve and
sedimentation PSD and these corrected PSD curves release true information about the sizes below 38
µm size.

Copyright © 2019 Isaac Scientific Publishing CWiMP

38 Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019

(a) (b)

Figure 4. (a) Uncorrected and (b) Corrected PSDs of ball and rod milled quartz by combination of PSDs measured
by sieving and Andreasen pipette sedimentation techniques (Redrawn from [21])

Table 5. values of d50 and apparent mean shape factor (r) determined by 2½ order size fraction (-53+38 m) for
quartz minerals measured by Sympatec laser diffraction and Andreasen pipette sedimentation techniques,
respectively [21, 22].

Method Mill product d50 (m) Apparen apparent mean shape factor (r) or (1/r)
laser diffraction Ball 34.7 1.29 (r)
Rod 36.0 1.25 (r)
sedimentation Ball 35 1.00*(1/r)
Rod 40 1.12 (1/r)
* assumed to be unity since no corrections needed in the PSD see (Figure 4 (a))

(a) (b)

Figure 5. SEM microphotographs for (a) ball and (b) rod mill products of quartz (x100) (from [21, 22])

The SEM microphotographs about the different milled quartz products in Figure 5 gives also some
evidence related to the shape differences between the sedimentation and sieving data and r factors. in
fact, this work gives conclusion of rounder particles have produced by balled milling comparing to rod
milling products as good agreement with the previous studies [39, 35, 40], i.e., r value of 1.00 for ball
milled quartz was explained by the differences between the material structure of the minerals used, i.e.,
the material properties have as much marked effects as the mill type reported by [41, 42, 39, 35]. It

CWiMP Copyright © 2019 Isaac Scientific Publishing

Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019 39

should also be noted that, the Stokes equation theoretically prevails for only spherical particles. But, the
difference between the sedimentation and sieving PSD does not represent erroneousness on the contrary
releases a useful information for particle shape. The more irregular the particle, the smaller is the shape
factors. As shown by Figure 5 from the SEM microphotographs, the rod milled product of quartz
studied does not have more regular shape according to the particle shape of ball milled products of the
same mineral. This was attributed to the different breakage mechanism acted in the ball and rod milling.

4.2 Integration of PSDs by Laser Diffraction and Sieving Techniques

While Figure 6 (a) illustrates the coarse PSDs of by ball and rod mill products by using sieving
technique in Schuhmann form down to 38 µm size, fine PSDs below 38 µm by laser diffraction method
for the same samples ground in different mills were represented by Figure 6 (b) indicating also a good
smooth size distribution. Both of them in deed, show small differences between the ball and rod milled
particles.

(a) (b)

Figure 6. PSDs of quartz minerals ground in ball and rod mill by (a) Sieving and (b) laser diffraction techniques
(Redrawn from [22])

(a) (b)

Figure 7. (a) Uncorrected and (b) Corrected PSDs of ball and rod milled quartz by combination of PSDs measured
by sieving and laser diffraction techniques (Redrawn from [22])

Copyright © 2019 Isaac Scientific Publishing CWiMP

40 Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019

For the integration of coarse and fine PSDs measured by different techniques, namely sieving and
laser diffraction techniques, the uncorrected PSDs on a log-linear scale of the same quartz samples
ground by different mills were produced by multiplying the cumulative undersize value of 38 µm for
sieving technique by the values of cumulative weight % undersize for laser diffraction technique as
depicted by Figure 7 (a). Even though PSD data were well presented Figure 7 (a) there were some
deviations between the fine and coarse PSDs by laser diffraction size distributions and sieve size
distributions. This could be useful information on particle shape of the same material by different
techniques [43, 44, 36, 2, 32, 33, 38].
The sieved fraction size of (geometric mean size fraction) -53+38 m for quartz mineral ground by
ball and rod mills was measured by laser diffraction sizing technique. Then, apparent mean shape
factors, r, were calculated based on the equation 2 once d50 values were found from the laser diffraction
tests for the 21/2 sieve size fraction (-53+38 m) as given in Table 5. Afterwards, the corrected PSDs
(Figure 7 (b) were obtained by utilization of apparent mean shape factor (multiplying of the particle
size with r factor), thereby shifting the sieving PSDs to the right for the rod milled barite and quartz
and ball milled quartz. Table 5 also shows the calculated apparent mean shape factors from the
corrected particle size distributions of quartz minerals ground by ball and rod mills. r values were found
as 1.29 and 1.25 for ball and rod mill products, respectively, i.e., both of them are nearly the same and
indicating that more deviation from the shape of sphere.

5 Conclusion
As it is too difficult to obtain a PSD of fine industrial particles by using sieving technique, which is not
so true for less than 38 m, the PSDs of fine particles by using another different techniques, such as
Andreasen pipette sedimentation and laser diffraction can be easy and the entire PSD including coarse
and fine PSD can be integrated by the combination of the two PSDs of the same material by these
different sizing techniques, i.e., fine PSD of sedimentation or laser diffraction and coarse PSD of sieving
by utilizing an approach of apparent mean shape factor, which also reveals the particle shape
information from the difference of the two PSDs for the same material by measuring them with different
sizing techniques. Thus, this approach can be utilized for not only quartz but also for the fine
particulate coals, minerals, and similar material ground finely.

References
1. Pabst W. & Gregorová, E., (2007). ICT Prague 2007, Characterization of particles and particle systems.
2. Allen, T. (1990). Particle Size Measurement, 4th ed., New York: Chapmann and Hall.
3. Jillavenkatesa, A., Dapkunas, S. J., & Lum, L. H., (2001). Particle Size Characterization, Special Publication
960-1, NIST Recommended Practice Guide, p. 27.
4. Horiba Instruments Inc , (2010). A Guidebook to particle size analysis
5. Stanford, R.E., & Patterson, B.R., (2007). Controlled particle size distributions using Linear Programming,
Powder Technology, 176, 114–122.
6. Cirulis, D., (2017). Particle Size Tracking System vs. Traditional Measurement Techniques, E&MJ, March, p.
58
7. Kelly, E.G. & Spottiswood, D.J. (1982). Introduction to mineral processing, John Wiley &Sons Inc., New York,
USA, pp. 21–22.
8. Wills, B.A. & Napier-Munn, T., (2006). Wills' Mineral Processing Technology, An Introduction to the Practical
Aspects of Ore Treatment and Mineral Recovery, 7th revised ed. Elsevier Publisher,
9. Ikechuks, G. A., (2011). The effects of particle size on the wettability of Akwuke coal using continuous flow
technique, Proceedings of the World Congress on Engineering and Computer Science, San Francisco, USA, vol.
2, pp. 237–247.
10. Kök, M.V., Özbas, E., Hicyilmaz, C., & Karacan, Ö., (1997). Effect of particle size on the thermal and
combustion properties of coal, Thermochim. Acta, 302, 125–130.
11. Mohns, C.A., (1997). The Effects of Particle Size on the Kinetics of Coal Froth Flotation Master's thesis
Queen's University, Ontario, Canada.

CWiMP Copyright © 2019 Isaac Scientific Publishing

Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019 41

12. Boylu, F., Dincer, H., & Ateşok, G., (2004). Effect of coal particle size distribution, volume fraction and rank on
the rheology of coal–water slurries, Fuel Process. Technol., 85, 241–250.
13. Brikci-Nigassa, M., Garbett, E.S., & Hedley, A.B., (1982). The effect of coal particle size on the performance of
a fluidised bed coal combustor, Prepr. Pap. Am. Chem. Soc. Div. Fuel Chem. (U. S.) 27.
14. Schneider, C. L., Neumann, R. & Souza, A. S. (2007). Determination of the distribution of size of irregularly
shaped particles from laser diffractometer measurements, Int. J. Miner. Process. 82, 30–40.
15. Zhang, Z. Yang, J. Ding, L. Zhao, Y., (2012). An improved estimation of coal particle mass using image analysis,
Powder Technol. 229, 178–184.
16. Ko, Y.-D. & Shang, H., (2011). Time delay neural network modeling for particle size in SAG mills, Powder
Technol. 205 250–262.
17. Eshel, G. Levy, G.J., Mingelgrin, U., & Singer, M.J., (2004). Critical evaluation of the use of laser diffraction for
particle-size distribution analysis, Soil Sci. Soc. Am. J. 68 (3) 736–743.
18. Ghasemy, A., Rahimi, E., & Malekzadeh, A., (2019). Introduction of a new method for determining the particle-
size distribution of fine-grained soils, Measurement, 132, 79–86.
19. Oja, M., Tuunila, R., 2000. The influence of comminution method to particle shape, Proceedings of the XXI Int.
Min. Proc. Cong., Rome, Italy, July 23–27, p. C4-64-70.
20. Johnston, R.G., & Reuter, J. M. (1993). Mineral uses in paint and their effect on quality, SME Annual Meeting,
Reno, Nevada, 15-18 February.
21. Ulusoy, U., Yekeler, M., Biçer, C. & Gülsoy, Z., (2006). Combination of the Particle Size Distributions of Some
Industrial Minerals Measured by Andreasen Pipette and Sieving Techniques, Particle & Particle Systems
Characterization, 23 (6) 448-456.
22. Ulusoy,U., Gulsoy, Ö. Y., Aydogan, N. A., & Yekeler, M., (2008). Combination of laser diffraction and sieve size
distribution by determining the mean shape factors, Particulate Science and Technology, 26, (2) 158-168.
23. Leschonski, K., (1979). Sieve analysis, the cinderella of particle size analysis methods, Powder Technology, 24,
115–124.
24. Allen, T. (2003). Powder Sampling and Particle Size Determination, Elsevier Science Ltd., pp. 208–250.
25. Rhodes, M. J. (2008). Introduction to particle technology, John Wiley & Sons Ltd., pp. 12–13.
26. Allen, T., (1992). Particle Size Measurement, Chapman & Hall. 4th Ed.
27. Igathinathane, C., Ulusoy, U., & Pordesimo, L.O., (2012). Comparison of particle size distribution of celestite
mineral by machine vision Volume approach and mechanical sieving, Powder Technology, 215-216C, 137-146.
28. Lambourne, R., (1993). Paint and Surface coatings–theory and practice; (Ed.), Ellis Horwood Ltd., ISBN 0-13-
030974-5PGk
29. Levoguer, C., (2013). Using laser diffraction to measure particle size and distribution, May/June MPR, pp. 11-
18, metal-powder.net
30. Ma, Z., H. G. Merkus, J. G. A. E de Smet, C. Heffels, & Scarlett, B., (2000). New developments in particle
characterization by laser diffraction: size and shape, Powder Technology, 111: 66-78.
31. Kippax, P., (2005). Measuring particle size using modern laser diffraction techniques, Paint&Coatings Industry,
August.
32. Austin, L. G., (1998). Conversion Factors to convert particle size distributions measured by one method to
those measured by another method, Part. Part. Syst. Charact. 15: 108-111.
33. Austin, L. G., and I. Shah. 1983. Powder Technol., 35: 271-278.
34. Ulusoy U., & Yekeler, M., (2004). Variation of critical surface tension for wetting of minerals with roughness
determined by Surtronic 3+ instrument, Int. J. of Miner. Process., 74, (1-4): 61-69.
35. Ulusoy, U., Yekeler, M. & Hiçyılmaz, C., (2003). Determination of the shape, morphological and wettability
properties of Quartz and their correlations, Minerals Engineering, 16, (10): 951-964.
36. Stokes, G. G., (1891). Mathematical and Physical Paper III, Cambridge University Press,
37. Sympatec Company, Clausthal-Zellerfeld, Germany.
http://www.sympatec.com/LaserDiffraction/LaserDiffraction.html
38. Austin, L. G., Tras, O., Dumm, T. F., & Koka, V. R., (1998). Part. Part. Syst. Charact. 5, 13-15.
39. Kaya, E. Hogg, R. Kumar, S. R., (2002). Particle shape modification in comminution, KONA, 20, pp. 188-195.
40. Austin, L. G., Yekeler, M., Dumm, T. F., & Hogg, R., (1990). The kinetics and shape factors of ultrafine dry
grinding in a laboratory tumbling ball mill, Part. Syst. Charact., 7, 4, pp. 242-247.
41. Bond, F. C., (1954). Control particle shape and size, Chem. Eng., pp. 1028-1032.
42. Heywood, H., (1961). Powders in Industry, Soc. Chem. Ind., pp. 25-26.

Copyright © 2019 Isaac Scientific Publishing CWiMP

42 Current Works in Mineral Processing, Vol. 1, No. 1, March, 2019

43. Durney, T. E., & Meloy, T. P., (1986). Particle shape effects due to crushing method and size, Int. J Miner.
Proc., 16, 109-123.
44. Lefond, S. J, (1983). Industrial Minerals and Rocks, 5th ed., vol. 2, SME, Littleton,

CWiMP Copyright © 2019 Isaac Scientific Publishing